Amazon Rufus Agentic: What Brands Must Know

Fact: Amazon’s Rufus AI shopping assistant reached 115% year-over-year monthly active user growth and drove $12 billion in incremental annualized sales — confirmed on Amazon’s Q1 2026 earnings call by CEO Andy Jassy.
Impact: Rufus now places orders autonomously through Scheduled Actions — recurring reorders, price-trigger purchases, and gift reminders — without a shopper ever initiating a session.
Surprise: Customers using Rufus are 60% more likely to complete a purchase. This is not a search optimization problem. It is a product data quality problem.

There is a number buried in Amazon’s Q1 2026 earnings call that deserves more attention than the stock tickers. Monthly active users for Rufus — Amazon’s AI shopping assistant — grew 115% year over year. Engagement jumped 400%. In 2025, 300 million customers used it. And with the Scheduled Actions rollout now live, Rufus can buy products on a shopper’s behalf without them ever opening the app.

That shift — from AI that answers questions to AI that places orders — happened faster than most brand managers planned for. The question is no longer whether Rufus matters to your Amazon strategy. It is whether your product listings contain the machine-readable data Rufus needs to recommend them at all.

The Q1 Numbers That Should Change Your Catalog Priorities

CEO Andy Jassy described Rufus’s trajectory on the April 29 call with the confidence usually reserved for proven cash engines. Customers who interact with Rufus are 60% more likely to complete a purchase — not browse, but buy. Nearly $12 billion in incremental annualized sales have been attributed to the assistant. That is a number that does not exist unless Rufus is actively redirecting purchase decisions rather than merely influencing them at the margins.

What is striking about how those conversions happen is that Rufus does not optimize for keyword density. It answers natural-language queries by parsing structured attributes: item type, material, intended use, technical specifications, compatibility. A waterproof jacket engineered for 2022 keyword rankings can be invisible to a query like “women’s hiking jacket under €75 that ships in two days.” That query requires machine-readable answers to at least four distinct attribute questions. Most Amazon listings don’t have them.

Scheduled Actions: When Rufus Shops Without Being Asked

The Scheduled Actions announcement changes the game in a way the 115% headline does not fully capture. A shopper sets a rule: reorder protein powder every six weeks, buy a birthday gift before a specific date, purchase the moment a price drops below a target. Rufus sets the rule, monitors it, and executes — no search, no scroll, no checkout friction involved.

The new agentic capabilities include:

Recurring reorders — calendar-driven auto-replenishment without a user prompt
Price-trigger auto-buy — purchase fires automatically when a product hits a pre-set threshold
Shop Direct — cross-web buying that extends beyond Amazon’s own marketplace
Custom Shopping Guides — personalized product selections generated on demand

What we’re seeing at Epinium is that this removes brand visibility from the actual purchase moment. The decision is made earlier — when the shopper sets the rule. If your product was not what Rufus recommended at that moment, you are out of every future auto-buy that rule triggers. Permanently, unless the shopper actively revisits the decision.

Epinium data

Across Epinium’s managed catalog base — spanning more than 400 brand accounts on Amazon Europe and North America — fewer than one in four product listings currently carries the structured attribute depth that tools like Rufus need to surface a confident purchase recommendation. That gap is exactly where revenue is starting to leak.

Is your Amazon catalog Rufus-ready? Epinium’s catalog management platform audits and completes attribute data at scale →

Beyond Amazon: Why the UCP Move Signals a Structural Shift

Six days before the Q1 call, Amazon quietly joined the Universal Commerce Protocol tech council alongside Meta, Microsoft, Salesforce, and Stripe. The UCP is Google’s open standard for AI-driven shopping, launched in January 2026. Amazon notably sat out the original announcement — analysts at the time flagged it as a potential threat to marketplace dominance. The reversal signals that Amazon recognizes agentic shopping will increasingly route through multiple systems: Google’s AI Mode, OpenAI’s agents, third-party voice interfaces. Sitting out a standard means ceding influence over how those systems process product data.

As Epinium explored in AI Agents Can Now Browse Your Product Catalog, the structured data requirements for Rufus and for external AI agents converge on the same fundamentals: complete, machine-readable attribute sets. Brands that treat Amazon catalog optimization as an Amazon-only concern are already behind the curve.

Frequently Asked Questions

Is Amazon Rufus available in European marketplaces, or only in the US?

Rufus has rolled out across the UK, Germany, France, Spain, Italy, and several other markets throughout 2025 and 2026. Auto-buy and Scheduled Actions launched in the US first, but the attribute-driven recommendation logic applies globally. European brands should plan for full feature parity within 12 months and treat catalog completeness as urgent regardless of market.

How does Rufus balance sponsored placements against organic product results?

Amazon confirmed in its Q1 2026 coverage that Sponsored Products and Brand Prompts are now integrated into Rufus responses. Nearly 20% of shoppers who interact with a Brand Prompt continue the conversation about that brand — making Rufus an advertising surface as well as an organic discovery channel. Brands that ignore the paid integration risk losing the initial Rufus interaction to a better-funded competitor.

My brand is in a niche category with low Rufus query volume. Does this still matter?

It matters more than it might appear. The 60% purchase-completion lift means every Rufus interaction in any category carries outsized conversion weight relative to traditional search clicks. A specialist brand capturing even three Rufus-influenced conversions per day is accumulating revenue that keyword-only optimization would never have reached. The argument to deprioritize Rufus in tight niches does not survive contact with the math.

Can I use Rufus as a competitive intelligence tool against my own category?

Yes — and this is among the most underused tactics in Amazon catalog management right now. Query Rufus with your own product’s use cases and observe which competitor listings it recommends, and on what basis. The alternatives it surfaces reveal exactly which structured attributes your listing is missing. Run this audit before every major catalog review cycle.

How much catalog cleanup work is actually needed before Rufus impact improves?

In most categories, the critical attributes concentrate in 8 to 12 fields: item type, material, intended use, target audience, compatibility, and two to four category-specific technical specs. Completing those fields to Amazon’s recommended depth — not simply filling them with placeholder values — is typically a two-to-four-week sprint for a mid-size brand. Given the 60% purchase-completion lift, the ROI case is not difficult to make.

Amazon’s $181.5 billion Q1 result — 17% year-over-year growth at that scale — is partly a Rufus story. The Scheduled Actions rollout effectively turns the assistant into an autonomous purchasing agent for hundreds of millions of shoppers. That is not a feature update. It is a channel shift.

Brands that treat their Amazon catalog as a static asset will find Rufus routing around them. Those that build catalog quality as a dynamic, AI-readable data layer will find themselves recommended on autopilot — including in purchases that were decided weeks before the order was placed.

Ready to make your catalog Rufus-ready? Epinium’s platform audits attribute completeness across your entire Amazon catalog, identifies the specific gaps costing you Rufus recommendations, and resolves them at scale — in every language and marketplace. Discover how Epinium optimizes your Amazon catalog →